Empirically-driven hierarchical classification of stock keeping units

This paper proposes a hierarchical multi-criteria classification method developed for inventory management purposes and applied in a case study of the spare parts business of a household appliance manufacturer. The classification method is built on the basis of SIX dimensions, resulting in 12 different classes of spare parts, for which differentiated forecasting and inventory policies are proposed and tested. The results of our simulation study demonstrate the reduction of the total logistics costs by about 20% whilst still achieving the specified target service level for each class. Even more importantly, the proposed approach is simple enough to be understood and applied by company managers, thus increasing the probability of its adoption (in the same or similar fashion) in other real world settings.